processes after metatalk tasks
Rejane Teixeira Vidal
Universidade Federal Fluminense - UFF
Esta investigação combinou a instrução-focada-na-forma com o ensino baseado em tarefas em contexto de aprendizagem autônoma no cenário de ensino de inglês como língua estrangeira que tem por objetivo o desenvolvimento da interlíngua do aprendiz no que tange à precisão lingüística. Pretendeu-se dar uma contribuição ao entendimento de que, ao produzir a língua alvo enquanto se reflete sobre ela, o aprendiz pode consolidar conhecimento já existente assim como gerar conhecimento do que é novo para ele (SWAIN, 1998), ajudando-o a compreender e a tornar-se mais consciente de seu processo de aprendizagem como um todo. O estudo atingiu seus principais objetivos: trouxe evidências adicionais à reivindicação de que aprendizes eficientes compreendem o processo subjacente à sua aprendizagem e que tarefas com foco na forma que exploram o diálogo colaborativo via metafala têm o potencial de esclarecer questões do campo de estudo conhecido como “desenvolvimento do aprendiz” (BENSON, 2001), além de melhorar a qualidade de sua produção lingüística.
Introduction
In the past decade or so, the belief that some type of focus on the formal properties of the target language as being beneficial for learners to acquire it has been renewed. For this reason, the ultimate aim of the investigation was to explore ways which may help English as a foreign language (EFL) learners to enhance their target-language linguistic precision from a focus on form perspective. Three minor objectives were also pursued: 1) how different form-focused instructional tasks affected the written production of EFL learners; 2) how students, individually or in dyads, dealt with the way in which they were expressing their intended meaning when led to reflect on the language they were producing across tasks; and (3) how students perceived themselves going about their learning process.
This paper is a report of some findings of the broader research project (VIDAL, 2003) and focuses mainly on the results which provide additional evidence for the claim that successful learners understand the process underlying their own learning.
The area of SLA research which was the focus of the study was instructed SLA – specifically as it relates to the provision of form-focused instruction (FFI). Broadly speaking, the term includes both a focus on forms (i.e. predetermined focus on discrete items as in the long-established traditional way), and a focus on form (i.e. focus on the formal properties of the target language within the context of meaningful communication, either in unpremeditated (LONG, 1991) or in planned ways).
learning in a foreign language scenario would have a contribution to make to the area of second language acquisition research and teaching.
Alongside a growing interest in how L21 learners’ attention could be directed to form within communicative settings, “emphasis has also recently been on increasing learner autonomy in the learning process and on learner-centred approaches to both learning and teaching” (WILLIAMS, 1999, p. 305). In fact, considerable ground has been gained in favour of a learner-centred approach as alternative to a teacher-learner-centred approach in language classrooms. Such a shift in focus involves learners more actively in the learning process, giving learners more responsibility in their struggle to learn an L2 in addition to offering them opportunities to negotiate meaning and form. This shift in focus has given prominence to the model of instruction based on tasks2 (LONG, 1985; NUNAN, 1989; LONG; CROOKES, 1993; BYGATE; SKEHAN; SWAIN, 2001).
Thus, the principal aim of the main study was to explore ways to help EFL learners to improve their language accuracy via metatalk tasks – tasks in which there is talk about language use – by means of collaborative dialogue. Besides, the investigation, supported by sociocultural theory (VYGOTSKY, 1978), a theory which advocates that all higher forms of learning derive from social interaction, intended to discover how much learners could do by themselves – either working alone or collaboratively – without help from external sources such as the teacher, dictionaries or TEFL books in general. In addressing these multiple perspectives in a study with adult advanced learners of English as a foreign language in Brazil, I expected to be able make a contribution to pedagogical practices not only in the Brazilian EFL context but also in the field of L2 language teaching and learning in other linguistic and cultural domains which have some similarity to ours. Since in this paper I will put an emphasis on how learners perceive themselves going about their learning processes, only the literature on successful-learners’ studies will be succinctly reviewed.
Successful-learners’ studies
This section is meant to review some studies on successful learners and their potential for learning and for controlling their own learning. Research in this area indicates the growing motivation for the dissemination of the concept of autonomy – the ability of learners to control their own learning (BENSON, 2001).
The psychological argument in favour of autonomy in language education is that
we learn better when we are in charge of our own learning. Learning is more meaningful, more permanent, more focused on the processes and schemata of the individual when the individual is in charge. Being in charge may also increase motivation and a motivated learner is often a successful learner (CRABBE, 1993, p. 443).
Naiman et al. (1978, 1996), one of the most influential research studies on second language learning, set the ground for studies which investigated the language learner and his or her potential for learning. The research was based on the premise that “all forms of language teaching could be greatly improved if we had a better understanding of the language learner and of the language learning process itself” (NAIMAN et al., 1996, p. 1).
Naiman et al. ’s study on the good language learner contributed to a redefinition of the role of the teacher giving impetus to a number of subsequent studies on specific learning and communication strategies as well as on how learners might be trained in such strategies so that they could become better learners and more ready for autonomy (O’ MALLEY; CHAMOT, 1990; WENDEN, 1991; COHEN, 1998).
Rivers (2001) reported on a study, advocating “autonomy at all costs”, which analyzed self-directed language learning behaviours of experienced (about 20.000 hours of experience in language learning) adult Georgian/ Kazakh language learners at the University of Maryland at College Park based on qualitative data: extensive written work (questionnaires), self-reported data. Drawing on the notion of metacognition – an ability said to be used by successful learners – more specifically, on metacognitive self-assessment, i.e. “the ability to assess one’s own cognition” (RIVERS, 2001, p. 279), and also on metacognitive self-management, i.e. “the ability to manage one’s further cognitive development” (RIVERS, 2001, p. 279-280), the researcher sought support for the accurate use of metacognitive strategies to control the language learning process in autonomous learning. He agreed with the widespread claim that learners with better self-monitoring abilities perform better in self-regulated language learning while learners with poor monitoring skills are less able to manage their learning, and perform worse, than good monitors.
Results indicated that all learners assessed their progress, learning styles, strategy preferences, and any conflicts with teaching styles and with the behaviour of other learners regularly. The researcher concluded that “the accurate use of metacognitive, affective, and social strategies to control the language learning process and the learning environment is the hallmark of self-directed language learning” (RIVERS, 2001, p. 287).
Vidal (2002) reported on a small-scale study, with eight adult English-Portuguese majors from a Brazilian southeastern university, designed to explore the full range of possible language learning strategies across skills in relation to learning achievement, with an emphasis on the writing skill. Reported frequency of language learning strategy use was correlated with actual strategy use and ratings of written performance on the completion of consciousness-raising communicative tasks.
et al., 1978; Green and Oxford, cited in Cohen, 1998, suggested that higher-proficiency learners use more strategies than lower-higher-proficiency ones. Others, for example Chen, 1990, also cited in Cohen, 1998, indicated just the opposite. Similarly, there are studies which described strategies used by more effective language learners (O’MALLEY; CHAMOT, 1990) and correlated their use to positive learning outcomes. Besides, there are others who claimed that there are “good” learning strategies (RUBIN, 1975; PAIVA, 1997).
Detailed results will not be reported here, but suffice it to say that some research findings have been confirmed and some more insights gained in relation to language learning/use strategies: students of higher proficiency levels reported they used more metacognitive strategies and less memory and affective strategies and in fact they did so. Cognitive strategies were also reported to be used and were actually used. Social strategies were reported to be usually used as well. However, due to research design, social strategies could not be accounted for in the verbal protocols (think-aloud) to make it possible to confirm their actual use. Concerning the relationship between the range of strategy use across skills as well as actual strategy use as identified through the verbal protocols and ratings of written performance on writing tasks, achievement did not seem to be closely and/or only associated with high-rating scores. In view of the results, other variables, besides learners’ efficiency, might have to be taken into consideration. Overall, however, the use of language learner strategies (encompassing both language learning and language use strategies (COHEN, 1998) should be seen as important tools learners can make use of to control and improve their learning effort, since they are keys to both greater autonomy and more meaningful learning (OXFORD, 1990).
The growing impetus for research in autonomy in language learning has provided interest in learner-based approaches directly focusing on the production of behavioural and psychological changes that would enable learners to take greater control over their learning. Current approaches have emerged from two separate traditions: North American work on learning strategies and strategy training as well as European work on learner training. In the 1990s, the two approaches have tended to merge. A cover term – learner development – is being used by both traditions which do not maintain any clear distinction in approach any longer.
self-directed approaches in which learners are encouraged to train themselves through reflection on self-directed learning activities – could be related to the activities designed for the investigation I carried out. Having students reflect on their language use (through the think-aloud and pair-work procedures) along with their own evaluation of their learning process (questionnaire) should raise their consciousness towards their learning enabling them to take greater control over it. Such behaviour is considered a key to autonomy in language learning.
Research design
Since the main research aimed at exploring opportunities for pushed output/interlanguage stretching3 from a three-fold perspective – (1) how different form-focused instructional tasks affected the written production of EFL learners; (2) how students, individually or in dyads, dealt with the way in which they were expressing their intended meaning when led to reflect on the language they were producing across tasks; and (3) how students perceived themselves going about their learning process – more than one procedure to collect data was necessary: a written production activity, introspection, collaborative work, and a questionnaire.
To reach the first objective, three different ways of dealing with form via different task types was explored: a product, a process and a grammaticization approach (BATSTONE, 1994a, 1994b). These tasks illustrated the question of whether particular types of form-focused tasks would be more beneficial than others to language accuracy in a continuum of a more or a less explicit focus on form (FOTOS, 1994; SWAIN; LAPKIN, 1998, 2001a; STORCH, 1998; SKEHAN; FOSTER, 1999). The targeted item was the present perfect, a language feature that has always been considered a difficult one for Brazilian learners of English. The difficulty lies basically on meaning rather than form and it is often perceived as lack of accuracy and non-familiarity with the multiple uses of the present perfect. In the product task (Task 1), more explicit focus on form, learners were led to consider brief extracts and to reflect on the rules/use of the present perfect vs. past simple so as to write a small text
of their own creation in which they would make use of these verbs which were object of their attention in the input. In the process task (Task 2), less explicit focus on form, learners were implicitly led to construct a text from a picture and an outline which suggested the use of the present perfect by means of a few questions. In the grammaticization task (Task 3), the implicit/ explicit continuum was midway between the process and product task. Students were provided with words and parts of words along with pictures which constituted the input for their written text but which did not tell the complete story. They should make use of all the words and affixes and of the pictures in whichever order they wanted (APPENDIX 1).
To reach the second object, I dealt with negotiated interaction explored via conscious reflection on language use (SWAIN; LAPKIN, 1995; 1998, 2001, 2002, in press; KOWAL; SWAIN, 1994, 1997, SWAIN, 1998, 2001a, 2001b; WILLIAMS, 1999). Such a focus constituted the bulk of the major research (VIDAL, 2003). I examined how tasks that stimulated reflection on the target language affected language learning and production taking into consideration negotiated interaction by means of dialogue in dyads and auto-dialogue. This included a descriptive account of collaborative dialogue – solving-problem, knowledge building dialogue (SWAIN, 2000) – in two types of verbal protocols: pair work protocols (PW) and individual work (think-aloud) protocols (TA).
To reach the third objective, the main concern of this paper, I examined how learners perceived themselves going about their learning processes (NAIMAN et al., 1978; COHEN, 1998; COTTERALL; CRABBE, 1999; BENSON, 2001). The way that learners understand and evaluate their own learning as well as the learning opportunities at their disposal is thought to be related to their success and may also prove to be additional evidence in support of a certain type of form-focused instruction in task-based learner-centred contexts. I examined learners’ evaluation of their learning performance (BENSON, 2001) by having them answer a semi-structured questionnaire to determine whether they thought that a particular task type provided them with better opportunities to write a more accurate text than two others. In addition, they were asked about the data collecting procedures. According to Dickinson (1992, p. 31),
operate. This is true of rehearsal of language use in a real communication situation and of the performance of an exercise in the classroom.
The data collecting procedures that were applied could be summarized as follows: written work, oral work (dialogue with the self, concurrent with the written text, as well as dialogue with the other, after the first version of the each participant’s written text but about it) and a questionnaire.
The written work, i.e. the written text that would result from task completion, was submitted to grading criteria. For each task, students’ written production was rated (product evaluation) according to analytic and holistic grading criteria (adapted from HEDGE, 1988; OMAGGIO, 1986; along with SWAIN, 1985), (APPENDIX 2). The ratings on each task aimed at measuring task written performance/achievement across tasks. Such ratings provided the parameters for the evaluation of the effect of task type on students’ written performance.
The verbal protocols provided the data for the analysis of the collaborative dialogue, i.e. of the self-interaction as well as of the interaction in pairs. Such collaborative dialogues, audiotaped and transcribed, were analyzed in light of the features which called students most attention while engaged in language use: a slightly adapted and combined version of Kowal and Swain’s (1994) Language-Related Episodes (LREs); Types of Response (TR) (SWAIN, 1998), and Storch’s (1998) Types of Justification (TJ) for features of concern. Episodes which dealt with beyond linguistic aspects per se (BLREs) were also identified and coded. Both types of verbal protocols contributed with information about the opportunities students had for interlanguage stretching as they engaged in negotiated interaction to improve accuracy in their written production.
data” (NUNAN, 1992, p. 143), in this study, the questionnaire was analysed more qualitatively (APPENDIX 3).
In summary, the research design was meant to collect information that would make it possible for this researcher to examine both the product and the process derived from the learning activity.
In the next section, coding schemes and the questionnaire interpretation procedures are described.
Data analysis: coding schemes and questionnaire interpretation procedures
Owing to the fact that task completion involved the production of a written and a spoken text, two coding schemes were applied – one to rate the written texts and the other to code students’ collaborative dialogues.
The coding scheme for the written text is explained fully below since it was used as framework to make the link between written performance and achievement. The coding scheme for the collaborative dialogues (TA and PW) took into account LREs/ BLREs as already mentioned and because such data are not the focus of the present paper the scheme will not be described any further (see VIDAL, 2004; in press).
The written texts were rated analytically (AS) and holistically (HS) (HEDGE, 1988; OMAGGIO, 1986) by three experienced EFL teachers as follow.
As for the analytic scoring students’ written production was rated according to a set of five-point (Excellent to Very Good, Good, Average, Fair to Poor, Poor) multi-trait scales especially designed to assess four aspects of the written language that the students produced: 1) content – namely, clarity, originality, logic and relevance (0-20); 2) control over vocabulary – namely, range and use of idioms, appropriateness and correct word forms (0-20); 3) acceptability of grammar – namely, tenses, agreement, use of preposition, article, pronoun, linkers, sentence structure and word order (0-50); 4) mechanics – punctuation, spelling and length (0-10).
The holistic scoring also took into consideration a set of five-point scales: Excellent to Very Good (90-100), Good (80-89), Average (60-79), Fair to Poor (40-59) and Poor (0-39) (APPENDIX 2).
Learners’ individual mean scores on each task, considering the analytic and the holistic evaluation, were translated into numbers from 01 to 100, and used in this study to measure overall written performance by task. In other words, students’ scores were used to measure the effect of task type on learners’ performance. In fact, students’ written texts were marked by the three raters and the average of each rater’s mean was used as the respondent’s final score. Grades between 80 and 100 were parameters for successful or effective learners.
Tables (1 and 2) show how students’ final grades were calculated. The first table illustrates Student 2’s grades in Task 1.
TABLE 1
Distribution of teachers’ ratings by student and task type:
Student Task Teacher Mean HS AS Content Vocabulary Grammar Mechanics
88 88,33 87,67 16,67 16,33 45,33 9,33
2 1 1 85,5 8 5 8 6 1 6 1 6 4 5 9
2 9 4 9 5 9 3 1 8 1 8 4 8 9
3 84,5 8 5 8 4 1 6 1 5 4 3 1 0
The second table illustrates Student 2’s grades in Task 1, in Task 2, and in Task 3 taking into consideration the mean of the grades the three raters had given Student 2 in the three tasks. Student 2’s overall grade is the mean of the grades he had been awarded in each task. Note that the grades Student 2 had in Task 1 are the result of the mean of the marks given her by the three raters. The same procedure was used to calculate S2 grades in the other tasks.
TABLE 2
Distribution of teachers’ ratings by student and task type:
Student Task Mean HS AS Content Vocabulary Grammar Mechanics
2 1 8 8 88,33 87,67 16,67 16,33 45,33 9,33 2 90,17 9 1 89,33 17,33 17,33 45,67 9
3 91,5 9 2 9 1 18,33 1 8 4 6 8,67
The analysis of the questionnaire consisted of interpretative work on the database as follows.
First, a comparison across tasks of students’ own evaluation of each task, in percentage terms (Qs 1-8), which took into account the following traits: interest (utility and pleasure), usefulness, enjoyment, best choice and opportunity for language accuracy (TAB. 6). Second, a correlation between students’ type of task preference and teachers’ rating scores (mean by task) so that students’ achievement by task and their own performance perception could be related. Then, a correlation between students’ evaluation of their performance, pedagogical preference (Qs 7, 8 and 15) and teachers’ ratings was again taken into account to see whether students were consistent throughout. Finally, their evaluation of how they viewed the think-aloud and pair work procedures (Qs. 9-14) was considered as there was interest in finding out students’ opinion about the two techniques.
The participants
Twenty (20) Brazilian university students, English-Portuguese majors, from a federal university in southeastern Brazil, selected after having passed the Michigan Test of English Language Proficiency4 (the multiple choice component) constituted the sample for this study. Because achievement across tasks would be measured, a proficiency test seemed to guarantee some homogeneity. They were paired off male and female with the exception of one pair composed of two females (otherwise the pairing was at random).
Data analysis
The analyses of learners’ written performance as well as their answers to the questionnaire are considered below.
Learners’ written performance: teachers’ ratings on writing tasks
As pointed out earlier, three teachers rated the students’ written production: the researcher herself, and two other experienced teachers who did not know the real focus of the investigation. There was agreement between them in their evaluation of students’ written performance. Statistical analysis of teachers’ ratings indicated that there was no divergence between Teacher 1’s, Teacher 2’s and Teacher 3’s scores taking the mean of AS (analytic score) and HS (holistic score) in each task into consideration (Task 1 (p=0,424); Task 2 (p=0,144) and Task 3 (p=0,914)).5,6 So their rating scores could be considered reliable.
As regards the effect of type of task on students’ written production, taking into consideration the overall mean (i.e. the mean of the holistic (HS) and analytic (AS) scores given by the three raters), statistically the three tasks are different (p=0,000) with evidence in the data that each student did better on Task 3, apart from Student 4 (cf. TAB. 3; GRAPHS 1 and 2). In fact, there was a highly significant difference between Tasks 1 and 3 (p=0,000), and between Tasks 2 and 3 (p=0,001). Task 1 and Task 2 revealed themselves to be similar (p=0,135), nevertheless. (The Friedman and Wilcoxon tests were used for paired analysis).
5 This research was conducted using the decimal system in which a comma (not a point) represents the decimal place. I have retained this usage for the sake of simplicity and convenience.
TABLE 3
Distribution of Teachers’ Ratings: overall mean by student, task type, and students’ final overall grade
Student Task 1 Task 2 Task 3 Overall Grade (T1+T2+T3)/3)
1 51,67 83,25 91,25 75,39
2 8 8 90,17 91,5 89,89
3 71,33 62,83 73,33 69,16
4 43,67 80,33 62,33 62,11
5 68,5 76,17 80,83 75,16
6 78,83 88,83 9 1 86,22,22
7 66,5 65,67 70,83 67,66
8 91,17 84,33 95,67 90,39
9 64,67 6 6 8 4 71,15
1 0 41,33 59,17 59,67 53,39
1 1 49,17 76,83 87,5 71,16
1 2 46,17 69,67 7 7 64,28
1 3 75,5 85,17 9 3 84,55
1 4 7 8 67,67 8 2 75,89
1 5 78,67 62,5 78,83 73,33
1 6 57,5 5 4 79,33 63,77
1 7 80,67 83,67 89,17 84,50
1 8 81,5 73,33 76,17 77
1 9 45,17 39,83 62,5 49,16
2 0 46,17 67,5 74,17 62,61 Final average
by task 65,21 71,85 80,00
The graph below (GRAPH 1) illustrates final average by task in a scale of 0 (zero) to 100 (one hundred).
The next graph (GRAPH 2) illustrates overall mean by student and task type. It shows that students performed better in Task 3.
GRAPH 2 - Distribution of Student’s scores by Task: Final Average
Besides, if one takes into consideration the frequency distribution of rating scores, (TAB. 4 and GRAPH 3), these finding are corroborated. There was a difference between students’ scores in Task 1, Task 2 and Task 3 as follows:
TABLE 4
Frequency distribution of students’ scores by task
Score Task 1 % Task 2 % Task 3 %
0-60 8 40% 3 15% 1 5%
60-80 8 40% 1 0 50% 9 45%
80-100 4 20% 7 35% 1 0 50%
GRAPH 3 - Frequency Distribution of Student’s scores by Score and Task
The results indicated that the number of students who were awarded higher rating scores increased in Task 2 and even more so in Task 3 as compared to Task 1. In other words, the students were rated as higher on Task 3, the grammaticisation task, than on the other two tasks, as expected: 50% of the students (n=10) were awarded grades between 80 and 100 on Task 3, 45% (n=9) were rated between 60 and 80, and only 5% (n=1) between 0 and 60. In Task 2, 35% (n=7) were rated between 80 and 100, 50% (n= 10) reached between 60 and 80, and three, 15% (n=3), were rated below 60. In Task 1, only four (4) students, 20%, were awarded grades above 80, and 40% (n=8) were rated between 60 and 80 and the same number of students between 0 and 60.
It was likely then that Task 3’s design provided students with better opportunities for extended output, giving support to the idea that grammaticisation tasks will enable learners to produce a better-written text than tasks which deal with grammar as product and/or as process. In addition to differences in terms of frequency distribution, the means calculated per task type: Task 1= 65,21, Task 2= 71,85 and Task 3= 80,00, revealed that there were differences between task scores overall – with higher achievement in favour of Task 3. The Friedman Test indicated that the difference was highly significant (p=0,000). Interestingly enough, average holistic scores (HS) nearly always matched average analytic scores (AS) (TAB. 5). This suggested that experienced teachers seemed to be able to evaluate their students’ production well whether or not analytic criteria were taken into consideration, a finding which may bring some insight to testing procedures (but outside the scope of this investigation).
Besides, when looking at task performance by subscale, apart from mechanics (p=0,178, i.e. no significant difference), with a mean percentage of 71,67% in Task 1, 76,75% in Task 2 and 81,58% in Task 3 – average percentage reaching 76,67%, the highest among all subscales, students performance was rated as higher on the grammar subscale in Task 3 (83,33%). In Task 1, the grammar subscale reached only a percentage of 69% for the mean and in Task 2 74,77%. In statistical terms, this means that there was not any difference in the grammar subscale between Task 1 and Task 2 (p=0,140), but there was a highly significant difference in the grammar subscale between Task 1 and Task 3 (p=0,000), and between Task 2 and Task 3 (p=0,002) (TAB. 5).
scores for the content (76%) and vocabulary (76,50%) subscales were higher in Task 3 than in Task 1 and in Task 2. This indicated there was a highly significant difference between Task 1 and Task 3 as far as content is concerned (p=0,004), and a significant difference between Task 2 and Task 3 (p=0,028) in the content subscale as well. On the other hand, Task 1 and Task 2 were similar as to content (p=0,162). As regards vocabulary, the same pattern was found: Task 1 and Task 2 are similar (p=0,286); Task 1 and Task 3 were different (p=0,001) and Task 2 and Task 3 were different (p=0,002), which was, in fact, a highly significant difference.
TABLE 5
Percentage distribution of teachers’ ratings: Overall mean by task
Task O. Mean HS AS Content Vocabulary Grammar Mechanics
Task 1 65,21 64,08 66,33 59,75 63,58 69,00 71,67
Task 2 71,85 71,32 72,38 69,42 67,17 74,77 76,75
Task 3 80,00 79,68 80,32 76,00 76,50 83,33 81,58
MEAN 68,39 69,08 75,70 76,67
Therefore, the results indicated that, in the perception of the raters, Task 3 (the grammaticization task/picture story task) contributed to more grammatically accurate written production, and that it also contributed to more control over vocabulary and quality of content than the others. This was perhaps due to the fact that in Task 3 students worked “with a kind of unfocused context, with the pictures and the words sketching out the briefest outline of a plot” (Cf. BATSTONE, 1994a, p. 107), so that how they developed their story was the result of their creativity in handling the words and parts of words given to them as well as their effort to make their interpretation of the story clear to everyone else.
According to Batstone, “grammaticisation tasks exploit the notion of context-gap”7 by means of which learners have to attend to both meaning and form more efficiently. “This kind of guided language use is what we want to encourage when we teach grammar as [grammaticization]” (BATSTONE, 1994a, p. 107). On the other hand, in tasks which explore
grammar teaching as product (Task 1), students may not be necessarily led into real-time language use; very often they are considered receptive activities instead. A product task might give the learner a sense of direction, but even though they help learners formulate hypotheses about the target language, they are said “to be confronting the learner with a conveyor belt of target forms which cannot, on the first encounter, be consciously ‘structured’ into the learners’ working models” (BATSTONE, 1994b, p. 225, emphasis his); whereas in tasks which explore formal instruction teaching as process (Task 2), learners might be led to develop proceduralization, i.e. to form and mentally store language routines through experience in language use, though sometimes this may lead to excessive freedom and targeted forms may not appear at all, or may not be used adequately.
The questionnaire and learners’ self-evaluation of their performance and data collecting instruments
TABLE 6
Percentage Distribution of Students’ replies on types of Tasks (From Questions 1-8)
TASK 1 TASK 2 TASK 3
80% 90% 90%
(n=16) (n=18) (n=18)
5% 5% 5%
(n=1) (n=1)
(Other=10% (Other=5% (Other=10% +/-+5%) (n=1) n=2)
10=55% 10,40% 10=45%
9=5% 9,5=5% 9=20%
8=20% 9=5% 8,5=10%
7=15% 8,5=5% 8=15%
5=5% 7=20% 7=5%
6=5% 7=5%
5=5%
10=30% 10=45% 10=45%
9=5% 9,5=5% 9=20%
8=30% 9=15% 8=10%
7=15% 8=25% 7=15%
5=10% 7=5% 6=10%
4=5% 6=5% 3=5%
10% 35% 45%
20% 15% 50%
In teresti n g (Q . 1, 3, 5) No t In teresti n g (Q . 1, 3, 5) Usef u ln ess (Q . 2a , 4a ,6a ) E n jo ym en t (Q . 2b , 4b , 6b ) L ik ed B est B etter tex t Questionnaire:
1. O que você achou da Tarefa 1. Foi interessante?
2. Avalie a Tarefa 1
a) pela sua utilidade* numa escala de 1 a 10: b) pelo prazer de fazê-la numa escala de 1
a 10:
3. O que você achou da Tafera 2? Foi interessante?
4. Avalie a Tarefa 2
a) pela sua utilidade numa escala de 1 a 10: b) pelo prazer de fazê-la numa escala de 1
a 10:
5. O que você achou da Tarefa 3? Foi interessante?
6. Avalie a Tarefa 3
a) pela sua utilidade numa escala de 1 a 10: b) pelo prazer de fazê-la numa escala de 1
a 10:
7. Tendo em vista as Tarefas 1, 2 e 3, qual você gostou mais? Justifique sua preferência. 8. Qual tarefa deu-lhe a impressão de que
você estava produzindo um texto mais bem escrito, mais elaborado?
Utilidade: no sentido de que a tarefa, em questão, oferece-lhe oportunidade de produzir um texto correto/adequado.
For Questions 8-15 see below.
Although the students also said they liked Task 2 (the process task) their evaluation of the usefulness of Task 3, in the sense that it had helped them to write a more accurate text as well and their enjoyment in doing it, clearly prevailed over the others. Task 1 (the product task) was the activity they liked the least and which in fact got the lowest scores (Cf. TAB. 3).
Some students showed a high degree of correlation between performance ratings (such as S2, S6, S8, S13) and awareness but others showed more contradictory results (such as S3, S4, S5, S10). Excerpts from the three best students (grades 80-100 by task) and from some three other students are included below.
The answers provided by S2 (88 in Task 1; 90,17 in Task 2 and 91,5 in Task 3), deserved being highlighted. Not underestimating learners’ intelligence and capacity, Student 2 exceeded all my expectations towards the evaluation of the tasks by a student. S2, without having been given any kind of instruction about the aim of the activities she had undertaken, was able to explain the principles behind the three different tasks very well. She noticed the implicit-explicit continuum across tasks and most importantly that Task 3, the mid-term in the continuum, helped practice both syntax and vocabulary in context creatively. In fact, her answers are copied in toto below:
Q1. (About Task 1) “Os trechos ajudam a “guiar o raciocínio”, dando, de certa forma, um “modelo” para você realizar a tarefa. Esta nos dá a oportunidade de revisar os tempos verbais de forma menos cansativa do que os usuais exercícios de repetição de gramática”. (S2) Q3. (About Task 2) “Tenho dificuldade para “pensar rápido”; as idéias
demoram a aparecer. Preciso de muito tempo para redigir o texto. Às vezes isso se torna “chato”. Gostei muito dessa tarefa. [making reference to Task 2]. Acho que a presença de uma figura é muito importante, pois desperta a curiosidade e “ativa” a imaginação. A forma “diálogo” também facilita, pois te dá maior liberdade. Entretanto, você não tem uma linha para seguir e se fixa mais no “contexto” do que na “forma”. Te dá liberdade demais”. (S2) Q5. (About Task 3) “Achei que é o “meio-termo”: tem a figura para “ativar”
sua imaginação, e dá uma espécie de “roteiro” da estrutura que deve ser seguida. E obriga a pessoa a revisar gramática e vocabulário”. (S2)
Q7. (Task preference) “A 3, com certeza. Pois, como expliquei, é mais equilibrada”. (S2)
The answers S2 gave to the other questions show clearly that, in her opinion, both PW and TA procedures are useful and valid as pedagogical tools helping interlanguage development.
In relation to pair work:
Q9. “Foi ótimo! A conversa ajudou muito, pois é através do ponto de vista de outras pessoas que a gente consegue refletir sobre o que escreve e pensa. O fato de já ter intimidade com a pessoa foi essencial...” Q10. “Sim. Você pára para pensar sobre o que produz. No meu caso, então,
ajudou muito. Pois admito que tenho medo de “me arriscar”, costumo repetir sempre as mesmas estruturas, o que impede que eu progrida mais”.
In relation to the think-aloud:
Q11. “Não, mas é difícil explicitar o pensamento, o que não significa que não é bom e útil”.
Q12. “Com certeza. Vou passar a fazer mais isso, para tentar tirar minhas dúvidas e resolver meus problemas”.
Q13. “Sim. Perceber onde tem dificuldades”. Q14. “Não”.
In relation to comparison across tasks and teaching options: Q15. “A 3 certamente. Pelos mesmos motivos já citados. É interessantíssima
e “puxa” o raciocínio”.
Obviously, not all students had the same type of perception about their learning processes, but, as a whole, their answers allowed this researcher to correlate high performance ratings and awareness positively. Below excerpts from S1, who despite not reaching overall mean above 80 (because of his grade in T1), was able to provide answers which showed correlation between performance and awareness:
Student 1 was aware that he did not produce a good text in T1. In fact, two of the raters marked it two (2) for content, a subscale under the analytic criteria, because he did not really write about his experiences and achievements. Although the connection between achievement and awareness seem to have been contradicted, S1’s low mark in T1 might have occurred accidentally because he did well in the other two tasks and showed he was able to evaluate himself appropriately across tasks. S1 also said Task 3 was the one he liked best “Porque esta tarefa exige mais do raciocínio, da criatividade do inglês do aluno.” As far as the work in pair, he did not see any disadvantage in this type of activity; on the contrary, he said he only saw advantages: “[s]im. Todas as vantagens! Troca de idéias e experiências, principalmente”. “Sempre que conversamos acerca do que produzimos, conseguimos nos ajudar a crescer”. In relation to the think-aloud procedure, on the other hand, he said “[f]oi divertido, muito embora confesse que achei que me atrapalhou, um pouco, ter que ficar falando”. (...) “O único aspecto negativo, para mim, foi me sentir como que atrapalhado pela minha própria voz, enquanto estava pensando.” The positive aspect, in his view, seemed to be that saying one’s thoughts aloud, helps one to notice one’s difficulties better: “sim, dessa forma conseguimos gravar melhor nossas dificuldades”. Such a view is in agreement with claim that reflection about language use contributes to L2 acquisition (SWAIN, 1998).
language use to be jointly solved: “[p]odemos trocar idéias. Percebi que tinhamos as mesmas dificuldades”; “esse tipo de atividade traz benefícios no aprendizado de inglês”.
The next excerpts show more contradictory results, but still provide some more evidence for the claim that successful students are more aware of their learning process than less successful ones.
Student 4, (Task 1= 43,67 (S4 did not understand the activity, wrote about the targeted item), Task 2= 80,33 and Task 3=62,33), one of the two students who did not do well in Task 3, said he liked it best of all “[p]orque essa tarefa me proporcionou maior possibilidade de construção de um vocabulário adequado ao texto e à situação”. In fact, he seemed to do fine in the vocabulary and grammar subscales, but, as for content, he did not do well in the perception of the raters. S4 surprisingly also liked T1,“oportuna e enriquecedora a proposta dos exercícios e a ligação do raciocínio que estabelecem entre si, habilitando o aluno a criar o seu próprio discurso”, the task S4 was rated very low (43,67). So, in these cases, performance and awareness did not seem to have a correlation. However, S4 liked T2 (written text for T2 was marked above 80) as well: “achei interessante pelo fato de explorar a interpretação visual do aluno para que o mesmo construa sua redação a partir de uma situação real construída por ele mesmo”. Regarding the methodological procedures, he did not see any disadvantage in either PW or TA, and stressed that “[a] troca de experiências, no caso, de idéias, contribuiu de modo fundamental para o desenvolvimento do meu inglês”, and “[a]o pensar alto, o aluno dialoga com si mesmo e se concentra com maior raciocínio” showing that even S4 saw the two procedures as offering possibilities for interlanguage stretching.
vocabulary problems. The think-aloud work was just “different”, and she did not observe any stretching in her English as a consequence of saying her thoughts aloud, contrary to the opinions of the majority.
Student 10 (Task 1=41,33; Task 2=59,17; Task 3=59,67. Overall mean=53,39, the second lowest overall mean) did not seem to have made a good evaluation of her learning process. In fact, if her grades were really taken into consideration as a parameter for being considered a successful student, she would not fit the category. She liked Task 2 best of all because she thought it gave her opportunity to use her own imagination to create a written text. But when she answered Q15, the last question, she said she would make use of Task 3 with her students but would not give them the words: “[a] [Tarefa] 3 mas não daria as palavras que ele precisa usar. O incetivava a usar a criatividade com muita liberdade e seu próprio vocabulário”, missing the main point of the activity, which purposefully intended to demonstrate that when learners are given words to start with they are provided with a good degree of regulation, making it possible for them to apply grammar to words creatively. For some researchers, the entire process of interlanguage development can be seen as a gradual movement from lexis to grammar instead of from grammar to lexis. In relation to working with a partner, she enjoyed being helped and having some doubts cleared up. S10, however, was one among few who did not see any advantage in the think-aloud technique and thought it did not help her develop her English further.
The correlation between performance/efficiency and awareness should be taken with caution, however, as there might be other variables which could have interfered (VIDAL, 2002). Nevertheless, notwithstanding the instances of inaccuracy identified in their production throughout task performance, it was noteworthy that the great majority favoured the grammaticization task as offering good opportunities for interlanguage stretching.
In relation to the collaborative dialogue, with the self or in dyads, most students thought the think-aloud procedure, in spite of being unusual to them, helped them pay attention to their output in ways that raised their consciousness towards aspects, which would, otherwise, have remained unnoticed. The pair work was seen as a totally positive technique as it helped them solve many of their language problems and also fostered their noticing some aspects which had not been seen as incorrect or inappropriate. On the whole, they recognized the positive influences of both think-aloud and pair work procedures. Indeed, I strongly tend to believe that both procedures should be considered beyond mere data collection techniques.
Concluding remarks
This investigation combined form-focused instruction with task-based learning in the context of autonomous learning in an English as a foreign language scenario with the aim of stretching learners’ interlanguage as related to language accuracy. It was mainly grounded on a sociocultural theory of mind (VYGOSTKY, 1978) and on the important role output plays in the context of second/foreign language acquisition (SWAIN, 1985, 1998) when emphasis is given to learners’ conscious reflection on language use.
The underlying important finding in the study was not that grammaticization tasks seem to be a good pedagogical tool for the development of written precision but the corroboration that metatalk tasks provide opportunities for language learning in individual and collaborative settings in a way that helps (at least advanced L2 learners) learners’ awareness-raising in relation to their learning process.
I believe that the study has achieved one of the major aims it had set out to accomplish. It has contributed to furthering the understanding concerning how producing the target language while reflecting on it may trigger cognitive processes that both consolidate existing knowledge and generate linguistic knowledge which is new to the learner, helping them understand and become more conscious of their learning process as a whole and, therefore, contributing to interlanguage development.
I would like to finish this paper with Benson’s assertion: “[t]he primary goal of all approaches to learner development is to help learners become ‘better’ language learners” (BENSON, 2001, p. 142). To a great extent, I believe form-focused tasks which explore collaborative dialogue via metatalk have the potential to shed some light in the field of learner development besides pushing learners in their output.
References
BATSTONE, R. Grammar. Oxford: OUP, 1994a.
BASTONE, R. Product and Process: Grammar in the second language classroom. In: BYGATE et al. Grammar and the language teacher. Hertfordshire, Prentice Hall, 1994b.
BENSON, P. Teaching and researching autonomy. Essex: Pearson Education, 2001.
DOUGHTY, C.; WILLIAMS, J. (Ed.). Focus on form in classroom second language acquisition. Cambridge: CUP, 1998.
ELLIS, R. Introduction: investigating form-focused instruction. Language Learning S1: Supplement 1, p. 1-46, 2001.
FOTOS, S. Integrating grammar instruction and communicative language use through grammar consciousness-raising tasks. TESOL Quarterly, v. 28, n. 2, p. 323-351, 1994.
HEDGE, T. Writing. Oxford: OUP, 1988.
KOWAL, M.; SWAIN, M. Using collaborative language production tasks to promote students’ language awareness. Language Awareness, v. 3, n. 2, p. 73-93, 1994. KOWAL, M.; SWAIN, M. From semantic to syntactic processing. How can we promote it in the immersion classroom? In: KEITH JOHNSON, R.; SWAIN, M. (Ed.) Immersion education: international perspectives. Cambridge: Cambridge University Press, 1997.
LONG, M. H. A role for instruction in second language acquisition: task-based language teaching. In: HYLTENSTAM, K.; PIENEMANN, M. (Ed.). Modelling and assessing second language acquisition. Clevedon: Multilingual Matters, 1985. p. 77-99.
LONG, M. H. Task, group and task-based interaction. Plenary address presented at the RELO Regional Seminar, Singapore. University of Hawaii working papers in ESL, v. 8, p. 1-26, Apr., 1989.
LONG, M. H. Focus on form: A design feature in language teaching methodology. In: BOT, K. de; GINSBERG, R.; KRAMSCH, C. (Ed.). Foreign Language Research in cross-cultural perspective. Amsterdam: John Benjamins, 1991. p. 39-52.
LONG, M. H.; CROOKES, G. (Ed.). Tasks in a pedagogical context. Multilingual Matters, Clevedon, Philadelphia, 1993.
NAIMAN et al. The good language learner. Toronto: OISE Press, 1978, 1996. NORRIS, J. M.; ORTEGA, L. Effectiveness of L2 instruction: a research synthesis and quantitative meta-analysis. Language Learning, v. 50, n. 3, p. 417-528, 2000. NUNAN, D. Designing tasks for the communicative classroom. Cambridge: CUP, 1989.
O’MALLEY, J. M.; CHAMOT, A. U. Learning strategies in second language acquisition. Cambridge: CUP, 1990.
OXFORD, R. Language learning strategies. Oxford: OUP, 1990.
PAIVA, V. L. M. de O. Estratégias individuais de aprendizagem de língua inglesa. Letras & Letras, Uberlândia, v. 14, n. 1, p. 73-88, 1997.
RIVERS, W. P. Autonomy at all costs: an ethnography of metacognitive self-assessment and self-management among experienced language learners. The Modern Language Journal, n. 85, p. 279-290, 2001.
RUBIN, J. Study of cognitive processes in second language learning. Applied Linguistics, n. 11, p. 117-131, 1981.
SPADA, N. Form-focussed instruction and second language acquisition: a review of classroom and laboratory research. State of the art article. Language Teaching, n. 30, p. 73-87, 1997.
STORCH, N. A classroom-based study: insights from a collaborative text reconstruction task. ELT Journal, v. 52, n. 4, p. 291-300, 1998.
SWAIN, M. Communicative Competence: some roles of comprehensible input and comprehensible output in its development. In: GASS, S.; MADDEN, C. (Ed.). Input in Second Language Acquisition. Rowley, Mass.: Newbury House, 1985. p.235-253.
SWAIN, M. Focus on form through conscious reflection. In: DOUGHTY, C.; WILLIAMS, J. (Ed.). Focus on Form in Classroom Second Language Acquisition. Cambridge: Cambridge University Press, 1998.
SWAIN, M. The output hypothesis and beyond: mediating acquisition through collaborative dialogue. In: LANTOLF, J. P. Sociocultural Theory and Second Language Learning (Ed.). Oxford: Oxford University Press, 2000. p. 97-114. SWAIN, M. Examining dialogue: another approach to content specification and to validating inferences drawn from test scores. Language Testing, v. 18, n. 3, p. 275-302, 2001a.
SWAIN, M. Integrating Language Content Teaching through Collaborative Tasks. The Modern Language Review, Special Issue, v. 58, n. (1, p. 44-63, 2001b. SWAIN, M.; LAPKIN, S. Problems in Output and the Cognitive Processes They Generate: a step towards second language learning. Applied Linguistics, v. 16, n. 3, p. 370-391, 1995.
SWAIN, M.; LAPKIN, S. Interaction and Second Language Learning: Two Adolescent French Immersion Students Working Together. The Modern Language Journal, v. 82, n. 3, p. 320-337, 1998.
SWAIN, M.; LAPKIN, S. Task-based second language learning: the uses of the first language. Language Teaching Research, v. 4, n. 3, p. 193-220, 2000. SWAIN, M.; LAPKIN, S. Focus on form through collaborative dialogue: exploring task effects. In: BYGATE, M.; SKEHAN, P.; SWAIN, M. (Ed.). Researching pedagogic tasks second language learning, teaching and testing, London: Pearson, 2001. p. 99-118.
SWAIN, M.; LAPKIN, S. Talking it Trough: two french immersion learners’ response to reformulation. International Journal of educational Research (Special issue on the role of interaction in instructed language), n. 37, p. 285-304, 2002.
SWAIN, M.; LAPKIN, S. “Oh, I get it now!” From production to comprehension in second language learning. In: BRINTON, D. M.; KAGAN, O. (Ed.). Heritage language acquisition: a new field emerging. Mahwh, NJ: Lawrence Erlbaum. (29 manuscript pages, in press).
VIDAL, R. T. Is there a correlation between reported language learning strategy use, actual strategy use and achievement? Linguagem e Ensino, Pelotas, Rio Grande do Sul, v. 5, n. 1, p. 43-73, 2002.
VIDAL, R. T. Interlanguage stretching: the role of consciousness-raising communicative tasks and metatalk. PhD Dissertation (unpublished). Niterói, UFF, 2003.
VIDAL, R. T. O diálogo colaborativo, individual ou em par, e o desenvolvimento da interlíngua em tarefas de metafala. Revista do ISAT, São Gonçalo, Rio de Janeiro, n. 3, p. 17-40, 2004.
VIDAL, R. T. Form-focused tasks, collaboration via metatalk and interlanguage stretching. In: LIMA, Marília dos Santos; GRAÇA, Rosa Maria de Oliveira (Org.). Intersecções pedagógicas no ensino de línguas estrangeiras: Brasil-Canadá. Porto Alegre: Editora da UFRGS. (no prelo). VYGOTSKY, L. S. Mind in society. Cambridge, Massachusetts: Harvard University Press, 1978.
WILLIAMS, J. Learner-Generated Attention to Form. Language Learning, v. 49, n. 4, p. 583-625, 1999.
Appendixes
APPENDIX 1
TASK TYPES (T1/T2/T3)
TASK 1 (explicit) TASK 2 (implicit) TASK 3 (in-between) PRODUCT PROCESS GRAMMATICIZATION (pre-specified forms and (deployment of grammar (apply grammar to words
their meanings) through self-expression) in language use – from lexis to grammar)
• S’s given (four) “sample” • S’s given a “sample” dialogue • S’s given a short list of words narratives and a picture illustrating a and part of words (affixes, • S’s asked to consider the use different situation pronouns, verbs, etc.) plus
of the simple past and the • S’s given questions which a series of pictures present perfect in the extracts may be used as outline to • S’s asked to write a story • S’s asked to write a similar help them write the story based on the pictures which
story on their “experiences (dialogue) they are supposed can be ordered in whatever and achievements”, to produce based order he wants using each using the simple past and on the picture word or part of word at present perfect as they are • S’s asked to write a dialogue least once
APPENDIX 2
GRADING CRITERIA
(Adapted from Tricia Hedge (1988) and Alice Omaggio (1986) 1 ANALYTIC SCORING
SCORING CRITERIA
CONTENT
Excellent to Very Good 20-18 Clarity
Good 17-16 Originality
Average 15-12 Logic
Fair to Poor 11-9 Relevance
Poor 8-0
VOCABULARY
Excellent to Very Good 20-18 Range/use of idioms
Good 17-16 Appropriateness
Average 15-12 Correct word forms
Fair to Poor 11-9
Poor 8-0
GRAMMAR/USAGE
Excellent to Very Good 50-46 Tenses
Good 45-40 Agreement
Average 39-30 Preposition
Fair to Poor 29-25 Article
Poor 24-0 Pronoun
Linkers Sentence structure
Word Order MECHANICS
Excellent to Very Good 1 0 Punctuation
Good 9 Spelling
Average 8-7 Length
Fair to Poor 6-5
Poor 4-0
2 HOLISTIC SCORING/OVERALL IMPRESSION
Excellent to Very Good: 90 – 100 (A) Good: 89 – 80 (B)
Average: 79 – 60 (C) Fair to Poor: 59 – 40 (D) Poor: 39 - 0 (F)
+/ - (cf. Swain, M., 1985):
Syntactic errors: relative to the number of finite verbs produced Verb errors: relative to the number of verb forms produced
APPENDIX 3
QUESTIONNAIRE
1. O que você achou da Tarefa 1. Foi interessante? 2. Avalie a Tarefa 1
a) pela sua utilidade* numa escala de 1 a 10: ——— b)pelo prazer de fazê-la numa escala de 1 a 10: ———-3. O que você achou da Tafera 2? Foi interessante? 4. Avalie a Tarefa 2
a) pela sua utilidade numa escala de 1 a 10: —b)pelo prazer de fazê-la numa escala de 1 a 10: ———-5. O que você achou da Tarefa 3? Foi interessante? 6. Avalie a Tarefa 3
a) pela sua utilidade numa escala de 1 a 10: ———— b)pelo prazer de fazê-la numa escala de 1 a 10:
————-7. Tendo em vista as Tarefas 1, 2 e 3, qual você gostou mais? Justifique sua preferência. 8. Qual tarefa deu-lhe a impressão de que você estava produzindo um texto mais bem
escrito, mais elaborado?
9. O trabalho colaborativo, em par, ajudou você no desenvolvimento do seu inglês? Contribui para esclarecer suas dúvidas?
10. Você viu/vê alguma vantagem/ponto positivo no trabalho em par? Qual? 11. Você viu/vê alguma desvantagem/ponto negativo no trabalho em par? Qual? 12. Pensar alto ajudou você no desenvolvimento do seu inglês?
13. Você viu/vê alguma vantagem/ponto positivo nesse procedimento de pensar alto enquanto produz um texto escrito? Qual?
14. Você viu/vê alguma desvantagem/ponto negativo no procedimento de pensar alto? 15. Se você fosse utilizar uma dessas tarefas com seus futuros alunos, qual você
escolheria? Por quê?
